Holley Super Sniper - Misfire at High Loads

Hello Everyone,

I've been lurking for a while but finally bit the bullet and joined HPA... I know this is my first post but I'm already in deep trouble... with a Holley Super Sniper throttle body EFI (8x100lb injectors) - inside a housing that looks like a Holley double pumper carb... It also controls timing through a Holley Hyperspark Dizzy (Hall effect sensor with traditional rotor and cap with 8 plug wires). The ECU is buried inside the throttle body too - and seems prone to RFI problems...

The engine is a 598 cu in Chevy big block - merlin block, tall deck. Dart aluminum heads, single plane air gap Dart intake, solid roller cam with 740 lift, 302/306 duration. Engine dynoed at 809HP on 93 pump gas.

I've literally been fighting to make this run for 6 weeks - and finally taking it to a chassis dyno shop on Wednesday... I can attach the tune file, but it won't mean anything to you unless you've worked with Holley sniper or terminators before. I've actually enlisted the help of a remote tuner with a good rep - and at this point, I don't think it's the tune... I think we have an ignition or internal engine problem that I can't diagnose.

I've been chasing RFI issues in this car since day 1 with the sniper. I finally have it stable enough to drive around and not randomly flame out with the hood closed (would run fine with the hood open). But... I've still got a severe misfire at 40-100%TPS during a road test - which is scary enough with this car (68 Camaro set up for drag racing). It also has a TCI 6x auto (converted 4L80E auto) and that's not helping things - shifting control isn't great. Hoping we can do WOT runs on the dyno and find the problem - before we kill someone driving on the streets.

New MSD superconductor wires and autolite spark plugs gapped at 40 thou. Plug color seems good if you don't let it idle around forever. I verified that the sniper and the timing light agree by locking timing at 20 degrees.

Will do another plug check and compression test before we go to the dyno - and hopefully have a look at the valve lash too. I don't want to waste tuner/dyno time...

Is there anything else you would suggest before we spend time/money at the dyno?

(And for reference, I'm the almost 60yr old guy running around a hot rod shop filled with 20 & 30 yr old metal fab and painter/bodywork types... who want nothing to do with laptops or tuning). They're probably ready to throw me and this Camaro into the scrap heap... Help !!!

Seems to be two coil variations, which one are you using?

Strange that it seems OK with the bonnet open, but has a problem with it closed - have you carefully checked the electrical systems - low and high voltage - for clearances, bearing in mind the engine will torque over under load? It seems like something is being disturbed by the bonnet, one thing may be the leads being fully pushed onto the spark plugs, and another thing to follow up on are the distances between the leads, especially 5&7, and if they're run close to a grounded surface.

Have you checked the phasing of the rotor to the cap terminals? While there isn't a mechanical or vacuum advance, the firing to rotor position is still going to change with RPM and load, and if it's out it could cause a miss-fire or cross-fire. IIRC, the BBC dissy rotates anti-clockwise, so the rotor should be aligned with the right side of the tip to the left of the terminal, so as the spark is advanced it sweeps to the right with the left tip to the right of the terminal.

I'd also drop the gap to 30 thou', especially with a higher compression engine.

If you can, borrow a 4/5 gas analyser - it makes it obvious is a high lambda is due to a mis-fire or a lean mixture, from the HC and CO values.

The Holley Sniper "family" is known to have RFI issues - with the old distributor-based ignition systems (on the older V8s that they are intended to be used on...) The prevailing theory is that the firewall can act as an antenna for the RFI created under the cap (and/or in the wires) and this corrupts the processor in the ECU brain. But... you see this in the data logs of RPM or TPS or the WBo2 - and we finally (after a bunch of grounding and shielding) don't see the noise in the data logs.

Regarding the rotor phasing, the Holley hyperspark comes with a rotor phasing alignment tool - and I followed all their instructions regarding install. As a final check - you can lock the commanded timing at 20 deg BTDC - and check with a timing light - and everything agrees. The small and big block chevy rotors turn clockwise.

It's funny... I keep hearing that Snipers are "prone to RFI", yet, I've installed and tuned dozens of them and not had any real issues with RFI and the only one that I did have was on my own car, that has no hood, or engine cover of any sort, and that was simply because I ran my crank wire (yes I'm using an MSD crank trigger with my Sniper), too close to spark plug wires. I had the spark plug wires running under the headers, and my crank sensor wire was within a few inches of them. The solution was to move the spark plug wires above the headers and the issue cleared right up. The issue was only noticeable at low RPM where there would be RPM spikes, high RPM didn't experience any noticeable issues.

I find that any issues that seem like RFI, are typically due to installation errors. Note, I'm not saying poorly done installs. By errors, I'm meaning like what I did on my own car, by running a high voltage source near a low voltage signal. We all want our wires ran nicely and sometimes this means we need to run wires together that shouldn't be. There is also the unavoidable need to cross wires, some that generate noise and some that don't like the radiated noise. The closer to 90 degrees you can cross these wires at, the higher the rejection to noise there will be. Also excessively long wires that are coiled up become antennas and can cause noise issues. When possible, shortening wires, such as those in adapter harnesses can help reject the possibility of noise being injected into signals. Other common issues are wrong spark plugs, or spark plug wires. I've seen people use non resistor types for either or both and this causes issues. On a personal note, using any MSD spark plug wire seems to just be an issue, they have been bad out of the box or wear very quickly (months) causing RFI issues (with any EFI system, and/or audio systems). Then there's just poor grounding of the chassis, engine, frame, and electronics. If your hood isn't specifically grounded running a ground strap or even just a wire to the hood can help avoid it becoming an antenna/transmitter for noise and reduce the chance of electrical noise in circuits.

I find high RPM misfires are typically caused by incorrect set ignition settings, most commonly dwell being too low, not getting enough charge into the coil to fire the the mixture with the increased cylinder pressures. To a lesser extent I've seen latency of the system can cause issues, because when there is latency in the electronics between the RPM signal source and the coil, it can delay the spark enough to cause a high RPM misfire that you generally won't see at low RPM.

How do you have the TPS signal sourced for the trans controller?

A datalog of the engine running through all conditions, including idle, cruise, part throttle accel and high load will help diagnose this too, beyond just the GCF.

Looking at your GCF, I see a few issues.

Needing the IAC at 100 to start suggests that your throttle plates are not set correctly, along with other starting parameters just don't look right.

You have a lot of timing in the timing table, 34 degrees at idle is far more than any engine will need. Which brings me to the next likely incorrect setting, the cam selection. You have "stock" selected, which by looking at cam specs is not correct. A small "stock" cam wouldn't need to idle at 1100 RPM, so I think you have a number of tune issues going on, besides any possible mechanical issues.

Your target AFRs are richer than I like for idle and cruise, but as a starting point aren't bad, a richer engine will idle easier, to a point, so it doesn't hurt to start richer and lean it out as you get the tune more dialed in.

There are other things I do differently with Holley, but these wont cause issues at this point, just things I do to avoid future problems.

Yes, I got the vacuum canister trick the wrong way round, it does move the backing plate ACW, but the shaft turns CW.

I see some good suggestions already involving closing up the gap a bit, and software changes, so I'll wait to hear how those work for you before piling on. Please keep us posted.

I will reply in a few minutes from laptop but here are a few pics from my cell phone.

Plugs pulled today & compression test was normal: 155/160 ish in all cylinders

photos show the aluminum channel wire looms I removed and the various shielding we’ve done to the dizzy and wiring.

Someone yesterday in an unrelated electrical discussion suggested ferrite beads on the coolant and tps sensor wires, since they on the 5 volt buss and may be receiving RFI that’s ending up on the processor buss.

A lot of this sounds like an engine that hasn't been tuned yet, so just to clarify, it hasn't been tuned at all yet...correct?

You can easily cause misfiring if fuel or timing are far enough away from optimal.

When it made 809 HP was that on an engine dyno, stated by the engine builder, controlled with a different ECU, wiring, etc.?

"find high RPM misfires are typically caused by incorrect set ignition settings, most commonly dwell being too low, not getting enough charge into the coil to fire the the mixture with the increased cylinder pressures. To a lesser extent I've seen latency of the system can cause issues, because when there is latency in the electronics between the RPM signal source and the coil, it can delay the spark enough to cause a high RPM misfire that you generally won't see at low RPM."

I'm using all the recommended Holley Hyperspark distributor in the tune: 57.5 reference angle and 100uSec inductive delay - there's also a dwell table I could play with, but I don't have it selected

"How do you have the TPS signal sourced for the trans controller?"

The TCI 6x trans controller has it's own TPS sensor - that sits on the Sniper throttle shaft - and provides TPS signal only for the trans - no intermixing of signals or sensors

"A datalog of the engine running through all conditions, including idle, cruise, part throttle accel and high load will help diagnose this too, beyond just the GCF."

Will see what I've got and post it - will be a few days old but I know I've got one that show the misfires/spikes in WBo2

Looking at your GCF, I see a few issues.

Needing the IAC at 100 to start suggests that your throttle plates are not set correctly, along with other starting parameters just don't look right.

The IAC is parked at 100 because this Super Sniper IAC system is too small to provide enough air at start-up for a 598 big block. This engine should probably have the 4500 Dominator sized Sniper with larger everything... lesson learned.

"You have a lot of timing in the timing table, 34 degrees at idle is far more than any engine will need. Which brings me to the next likely incorrect setting, the cam selection. You have "stock" selected, which by looking at cam specs is not correct. A small "stock" cam wouldn't need to idle at 1100 RPM, so I think you have a number of tune issues going on, besides any possible mechanical issues."

Do you have much experience with very large cam, solid roller big blocks? This is the timing they like - for best torque at idle - if you call what this thing does as idling - it is far from civil - 8-9 inches of manifold vacuum at best. Both the "remote" tuner and I have spent considerable time trying different spark and fuel - trying to get best stable idle. 1000-1100 rpm is as low as she goes... I'm not sure select "stock" vs "street/strip" vs "race" cam does much if you're tuning the AFR target table anyway - I think that's all it changes. I will play around with a few fresh Sniper files and see what changes. Some of the online tuners suggest starting with stock in any case, and the non-stock Holley defaults are much too rich.

"Your target AFRs are richer than I like for idle and cruise, but as a starting point aren't bad, a richer engine will idle easier, to a point, so it doesn't hurt to start richer and lean it out as you get the tune more dialed in."

I had it leaner based on input from the engine builder but the remote tuner richened everything up. Since the shop owner was paying him and not listening to me... he wins.

Regarding the tune, I've just been trying to get the thing to idle and putt around in granma mode with flaming out with random EFI spikes. It was originally dynoed with a carb and mech advance dizzy at the builder - he provided a 4500-6500 WOT dyno data pull with 38 degrees of timing and 14.5 down to 13.6 AFRs. The remote tuner thought his wide band needed a calibration - and I agree - that's about a point to 0.5 more AFR than I would be comfortable with at WOT.

We are going to a local chassis dyno/tuner tomorrow with the car - so we'll see if they can solve the RFI? misfire issue.

"there's also a dwell table I could play with, but I don't have it selected"

Surely dwell has to be configured somehow? Perhaps it's a fixed value rather than enabling the ability to control a 3d table worth of dwell values?

Improving grounding of vehicle wiring, and ensuring sufficient ground path for spark energy seems important at this stage, but it also sounds like until a tuner tries to tune it, the lack of tune itself could be causing all manner of issues.

Hopefully you'll have a great day on the dyno tomorrow!

"Surely dwell has to be configured somehow? Perhaps it's a fixed value rather than enabling the ability to control a 3d table worth of dwell values?"

Following the Holley guidelines for using an MSD Digital 6AL capacitive discharge ignition box - Dwell time is set to 2msec (plus the Reference angle and Inductive delay settings mentioned above)

"Improving grounding of vehicle wiring, and ensuring sufficient ground path for spark energy seems important at this stage, but it also sounds like until a tuner tries to tune it, the lack of tune itself could be causing all manner of issues."

Are you familiar with how the "self tuning" Sniper systems work? They have a base tune - and then the WBo2 is supposed to correct the long term fuel trims - which you can then copy over and modify the VE table. I think it actually works pretty well for the average street mild cam American 350 cu inch V8... but a 598 big block is another matter. That said it does actually drive ok now at less than 30% TPS - with the exception the horrible 6 speed 4L80E that attempts to get into 6th gear by 30 mph.

This car has more grounding straps than a radio tower - and Sniper power is run directly to the battery. Frame to body, engine to frame, engine to body, alternator to engine & frame. Hood hinge to firewall. Firewall to frame. Plus all the shielding shown in the pics above.

The only thing I haven't done yet is Ferrite beads - ordered those today. Will put them on all the 5 volt sensor lines - coolant and TPS for sure. Maybe WBo2 as well.

"Hopefully you'll have a great day on the dyno tomorrow!"

Thank you - we are due for some good luck on this car !!!

datalog attached

James, you're referring to how the system determines how much fuel to inject.

Coil dwell time is how long the coil charges per spark event. They are not related. Since you're getting misfires and one potential cause is insufficient spark, a few of us have been trying to help you determine if coil dwell control is set up properly.

Grounding sounds good, hopefully your dyno time goes well today!

“Coil dwell time is how long the coil charges per spark event. They are not related. Since you're getting misfires and one potential cause is insufficient spark, a few of us have been trying to help you determine if coil dwell control is set up properly.“

I guess I’m not following. The 2 msec setting above is the only dwell setting in the global file. There’s also nothing dwell adjustable on the msd 6a cds box. I do know the 6a box is only capable of multiple sparks below 3600 rpm, there’s no time above rpm for multiple sparks so it switches to one “allegedly powerful” spark

no word from the dyno tuner on the WOT runs, he may have run out of time. I was busy on other projects all day… I suspect I’ll be getting a call for help in the AM…

Ah gotcha. If using an msd box then you’re just triggering it with the ecu, not driving a coil, so static 2ms should be fine. I hope things went well!

Well well it seems the tuner is having the same high load/speed misfire issues we were having… he’s going down the “power and ground” everything to the battery road…which is where we were 4 weeks ago… so far the Sniper is winning and making all of us look like dummies…

Only new suggestion I’ve gotten this week is to put ferrite beads on all the 5v signal wires (coolant and TPS) with the theory that the 5v buss also powers the CPU…

Otherwise I’m fresh outta ideas…

Do you have a datalog of what happened on the dyno?

Not yet. Tuner and his tech called in sick today (the Devil-possessed Camaro will do that to you).

I do know he was seeing the same wild WBo2 swings we were seeing during our road test.

Hi

You metionned that you use MSD spark plug wire. Do you know if they are solid core or spiral core ? Solid core wire don't have any emi/rfi suspression and make ecu,ignition box goes crazy even making it bad.